Effect of hormone replacement therapy on cardiac ...

2 downloads 0 Views 205KB Size Report
Abstract. Objective To investigate the effects of hormone replacement therapy (HRT) on heart rate variability (HRV) in healthy postmenopausal women. Methods ...
Clin Auton Res DOI 10.1007/s10286-014-0226-1

RESEARCH ARTICLE

Effect of hormone replacement therapy on cardiac autonomic modulation Nata´lia Maria Perseguini • Anielle Cristhine de Medeiros Takahashi • Juliana Cristina Milan Patrı´cia Rehder dos Santos • Vale´ria Ferreira Camargo Neves • Audrey Borghi-Silva • Ester Silva • Nicola Montano • Alberto Porta • Aparecida Maria Catai



Received: 16 October 2013 / Accepted: 23 January 2014 Ó Springer-Verlag Berlin Heidelberg 2014

Abstract Objective To investigate the effects of hormone replacement therapy (HRT) on heart rate variability (HRV) in healthy postmenopausal women. Methods Two groups were evaluated: group 1 (G1): 20 women not undergoing HRT (60 ± 5.89 years), group 2 (G2): 20 women undergoing HRT (59 ± 5.70 years). The HRTs involved were either conjugated equine estrogen with or without medroxyprogesterone, synthetic estrogen hormone, estradiol associated with norethisterone acetate or isoflavonoids. Electrocardiogram was recorded in the supine position for 10 min. Spectral analysis included low

N. M. Perseguini  J. C. Milan  P. R. dos Santos  V. F. C. Neves  E. Silva  A. M. Catai (&) Cardiovascular Physical Therapy Laboratory, Nucleus of Research in Physical Exercise (NUPEF), Department of Physical Therapy, Federal University of Sa˜o Carlos, UFSCar, Via Washington Luiz, km 235, Sa˜o Carlos, SP 13565-905, Brazil e-mail: [email protected] A. C. de Medeiros Takahashi Geriatric Physical Therapy Laboratory, Department of Physical Therapy, Federal University of Sa˜o Carlos, UFSCar, Via Washington Luiz, km 235, Sa˜o Carlos, SP 13565-905, Brazil A. Borghi-Silva Cardiopulmonary Physical Therapy Laboratory, Nucleus of Research in Physical Exercise (NUPEF), Department of Physical Therapy, Federal University of Sa˜o Carlos, UFSCar, Via Washington Luiz, km 235, Sa˜o Carlos, SP 13565-905, Brazil N. Montano Department of Clinical Sciences L. Sacco, Internal Medicine II, L. Sacco Hospital, University of Milan, Milan, Italy A. Porta Department of Biomedical Sciences for Health, Galeazzi Orthopaedic Institute, University of Milan, Milan, Italy

and high frequencies in absolute (LF and HF) and normalized units (LFnu and HFnu), which are predominantly cardiac sympathetic modulation (CSM) and cardiac vagal modulation (CVM) indicators, respectively. The LF/HF ratio was also calculated. Symbolic analysis involved the following indexes: 0V % (CSM indicator), 1V % (CSM and CVM indicators), 2LV % (predominantly CVM indicator) and 2UV % (CVM indicator). Shannon and conditional entropies were also calculated. Results Spectral analysis demonstrated that HRT affected HRV. LF, LFnu and LF/HF ratio were higher (showing increased CSM), while HFnu was lower (representing decreased CVM) in G2 than in G1. Correlations between complexity indices and HFnu were significant and positive only in G1. Interpretation Women undergoing HRT presented higher CSM and lower CVM than those who were not. Moreover, the expected positive relationship between CVM and complexity of HRV was found only in control group, thus indicating that CVM in women under therapy drop below a minimal value necessary to the association to become apparent, suggesting an unfavorable cardiac autonomic modulation in spite of HRT. Keywords Heart rate variability  Symbolic analysis  Conditional entropy  Spectral analysis  Hormone replacement therapy

Introduction Heart rate variability (HRV) is an important approach for noninvasive analysis of cardiac autonomic function that refers to oscillations in the intervals between consecutive heartbeats, known as RR intervals (RRi) [1]. HRV has been

123

Clin Auton Res

widely used as a predictive factor, and its reduction is associated with higher cardiovascular morbidity and mortality rates [2]. HRV is commonly analyzed using linear models such as spectral analysis. However, interest in nonlinear methods has increased in recent years. Since the mechanisms that involve cardiovascular regulation are interconnected in a nonlinear way, nonlinear analysis could provide additional information. Nonlinear analysis differs from traditional approaches because it considers qualitative properties of heart rate (HR) time series [3–7]. For this reason, Porta et al. [8] developed a nonlinear methodology, symbolic analysis, that is applied in short-term recordings of RRi and has been effective for evaluating the sympathetic and vagal modulations of HR [9, 10]. Hormone replacement therapy (HRT) could influence cardiac autonomic modulation assessed by HRV. It has been reported that women undergoing HRT present greater HRV, with higher vagal modulation [11, 12] and lower sympathetic modulation [12, 13], than women not undergoing HRT. However, Neves et al. [14] found no differences in supine-position HRV between middle-aged women who were and were not undergoing HRT. On the other hand, these same authors observed that in sitting position, women undergoing HRT presented higher vagal modulation and lower sympathetic modulation than those who were not. Furthermore, they observed that nonHRT group had lower vagal modulation and higher sympathetic modulation than young women [14]. In addition, lower blood pressure [15, 16], decreased resting HR [16] and increased baroreflex sensitivity [15, 17] have been associated with HRT. Studies have shown that HRT could affect metabolism of lipids and lipoproteins [18] and atherosclerotic plaques formation [19]. Nevertheless, no studies have compared the cardiac autonomic modulation of women undergoing HRT or not by means of nonlinear methods such as symbolic analysis and HRV complexity. The hypothesis of the present study was that female HRT users would present higher cardiac autonomic modulation, with higher vagal modulation and lower sympathetic modulation. Therefore, the objective of the present study was to investigate, using linear and nonlinear methods, the HRT effect on the HRV.

Methods Subjects Forty healthy women aged 49–70 years were selected for participation of the study. The volunteers were divided into two groups according to HRT use: group 1 (G1 = control group) included 20 women not undergoing HRT

123

(60 ± 5.89 years) and group 2 (G2) included 20 women undergoing HRT (59 ± 5.70 years). The types of hormone, used by the volunteers in G2, included conjugated equine estrogen (0.625 mg) (seven women), conjugated equine estrogen (0.625 mg) associated with medroxyprogesterone (four women), isoflavonoids (60 mg) (four women), synthetic estrogen hormone (tibolone—2.5 mg) (three women) and estradiol (2 mg) associated with norethisterone acetate (1 mg) (two women). The subjects were using HRT to reduce menopause symptoms such as hot flashes, cold flashes, irritability, mood swings and depression. All subjects were considered healthy based on clinical and physical examinations, laboratory tests, a standard electrocardiogram (ECG) and a maximum exercise test conducted by a physician. All volunteers were also diagnosed as postmenopausal, characterized by amenorrhea for at least 1 year. None of the subjects presented abnormalities in the cardiovascular or respiratory systems. The ECG results for all volunteers were negative for myocardial ischemia and arrhythmia both at rest and during the maximum exercise test. Smokers, drinkers, users of illicit drugs or users of drugs that could interfere with the cardiorespiratory system, subjects with neurological, cardiovascular or respiratory disorders, diabetics and individuals with arterial hypertension were excluded from the study. Ethical aspects All volunteers were informed of the noninvasive procedures that would be performed during the study. After agreeing to participate in the study, all subjects signed an informed consent form. The study followed the Declaration of Helsinki guidelines and was approved by the Human Research Ethics Committee of the Federal University of Sa˜o Carlos, Brazil (protocol #174/2011). Experimental procedures All subjects were evaluated in the morning to respect the different responses due to circadian influence. The experiments were carried out in a climate-controlled room (21–24 °C) with a relative air humidity of 40–60 %. Subjects were instructed to avoid both caffeinated and alcoholic beverages as well as any strenuous exercise on the day before the test protocol. They were also instructed to have a light meal at least 2 h prior to the test. On the day of the experiment, the subjects were first interviewed and examined to determine if they were in good health, if they had slept properly the night before, and that the controlling conditions (HR and systemic blood pressure) were within normal range. Before the experiment, the volunteers were familiarized with the equipment and the experimental procedure to reduce anxiety.

Clin Auton Res

Experimental protocol The subjects remained at rest for 10 min in the supine position to stabilize cardiovascular parameters. After this, an ECG was then recorded (CM5 lead) for 10 min, using an interface with a bioamplifier for ECG signals (BioAmp Power Lab, ADInstruments, Australia) and a system for biological signal acquisition (Power Lab, ADInstruments, Australia). During the ECG recording, respiratory movements were simultaneously recorded with a respiratory belt (Marazza, Monza, Italy) attached to the chest. All signals were sampled at a frequency of 400 Hz. The subjects were instructed to breathe spontaneously throughout the procedure. At the beginning of the experiment, the blood pressure of all volunteers was measured using the auscultatory method. Data analysis RRi sequences n = 250 in length were selected for each subject. The length of greatest stability was chosen from the central region of the time series. The initial and final RRi sequences were discarded. The same sequence was used for both the linear and nonlinear analysis. The mean and variance of RRi were also calculated.

Briefly, this approach is based on classifying the full range of RRi sequences into six levels (from 0 to 5), transforming them into a sequence of integers (i.e., symbols). Patterns (sequences of three symbols) are constructed based on the sequence of symbols. The number of patterns is reduced by grouping all possible patterns into a small number of families. The pattern families are as follows: (1) 0V: patterns with no variation [three equal symbols, e.g., (2, 2, 2) or (4, 4, 4)], (2) 1V: patterns with one variation [two consecutive symbols are equal and the remaining symbol is different, e.g., (4, 2, 2) or (4, 4, 3)], (3) 2LV: patterns with two like variations [the three symbols form an ascending or descending ramp, e.g., (5, 4, 2) or (1, 3, 4)], and (4) 2UV: patterns with two unlike variations [the three symbols form a peak or a valley, e.g., (3, 5, 3) or (4, 1, 2)]. The rates of occurrence of these families (0V, 1V, 2LV and 2UV %) were evaluated in the present study. Previous studies involving pharmacological blockade [9] and autonomic tests [8, 10] have indicated that the 0V % and 2UV % indexes are the most powerful markers, inside the above mentioned set, to track modifications of cardiac autonomic modulation. 0V % and 2UV % indexes are representative of CSM and CVM, respectively. HRV complexity analysis: Shannon entropy and conditional entropy

Linear HRV analysis HRV spectral analysis was performed using an autoregressive model [20, 21] on previously selected RRi sequences. Two main spectral components were considered, i.e., low frequency (LF: from 0.04 to 0.15 Hz) and high frequency (HF: from 0.15 to 0.50 Hz). The spectral components are reported as absolute units (LF and HF) and normalized units (LFnu and HFnu) and LF/HF ratio [1]. Normalization consisted of dividing the power of a given spectral component (HF or LF) by the total power minus the power below 0.04 Hz and multiplying the ratio by 100 [20, 21]. The LF component expresses sympathetic and vagal modulations simultaneously, but this index best represents cardiac sympathetic modulation (CSM), mainly LFun. The HF component is an indicator of cardiac vagal modulation (CVM) [1]. The respiratory rate was confirmed to assure that it was in the frequency range included in the HF band. All volunteers were within this range, except for one woman undergoing HRT who was excluded from the study. Nonlinear HRV analysis Symbolic HRV analysis The symbolic analysis of HRV, applied in short-term recording of RRi, has been described by Porta et al. [8].

HRV complexity analysis was performed using Shannon entropy (SE) and conditional entropy (CE). Shannon entropy represents the complexity of pattern distribution: the index is large if the distribution is flat (all patterns are identically distributed) and the index is small if some patterns are more likely, while others are missing or infrequent [8]. Conditional entropy was calculated to quantify the amount of information carried by a new sample that cannot be obtained from a sequence of past values, i.e., the CE is related to complexity measures of the dynamic relationship between a pattern and the next RRi [8]. CE was modified to define corrected CE (CCE). As a function of past values, it was shown that the CCE: (1) remains constant in case of white noise, (2) reduces to zero in case of signs entirely predictable, and (3) exhibits a minimum value if repetitive patterns are embedded in noise. The minimum value for CCE was regarded as the complexity index (CI), which was expressed in natural numbers. This index was normalized by the SE of the RR series to obtain the normalized CI (NCI), which allowed the expression of complexity in dimensionless units. NCI ranges from 0 (null information) to 1 (maximum information). The larger the CI and NCI are, the higher the complexity, and the lower the regularity of the series [22].

123

Clin Auton Res Table 1 Age, anthropometric and clinical characteristics of each volunteers group Characteristics

Women not undergoing HRT (n = 20)

Women undergoing HRT (n = 20)

Age (years)

60 ± 5.89

59 ± 5.70

Weight (kg)

62.86 ± 9.35

61.50 ± 7.98

Height (m)

1.58 ± 0.07

1.57 ± 0.05

25.23 ± 3.23

25.01 ± 3.21

11 ± 5.51

9 ± 5.54

BMI (kg/m2) Menopause time (years) HRT time (years)



7 ± 4.61

SBP (mmHg)

119.68 ± 10.64

116.50 ± 10.77

DBP (mmHg)

73.68 ± 8.51

77.50 ± 7.86

MET (unidade)

8.48 ± 2.18

9.91 ± 3.96

89.69 ± 10.18

88.91 ± 9.94

Triglyceride (mg/dL)

101.25 ± 34.11

136.07 ± 81.29

Total cholesterol (mg/dL)

199.13 ± 44.97

219.73 ± 50.05

LDL cholesterol (mg/dL)

114.00 ± 44.97

131.60 ± 38.68

HDL cholesterol (mg/dL)

61.53 ± 15.69

60.60 ± 11.09

Glycemia (mg/dL)

Data are expressed as mean ± SD BMI body mass index, DBP diastolic blood pressure, HDL high density lipid, HRT hormone replacement therapy, LDL low density lipid, SBP systolic blood pressure, MET metabolic equivalent obtained in maximum exercise test

Statistical analysis

Table 2 Heart rate variability as determined by spectral and symbolic analysis, as well as Shannon and conditional entropies Women not undergoing HRT (n = 20)

Women undergoing HRT (n = 20)

p value

Linear analysis RRi mean (ms)

880.40 ± 68.73

942.40 ± 98.26

0.026

RRi variance (ms2)

892.50 ± 625.43

1,026.41 ± 816.14

0.508

140.05 ± 201.48

260.06 ± 264.49

0.004

26.38 ± 23.93

50.87 ± 22.88

0.002

410.54 ± 369.01

349.63 ± 691.09

0.156

71.02 ± 23.17

47.66 ± 22.98

0.003

0.75 ± 1.40

1.89 ± 2.10

0.002

3.67 ± 0.46

3.51 ± 0.43

0.253

Spectral analysis LF (ms) LFnu (nu) HF (ms) HFnu (nu) LF/HF ratio Nonlinear analysis Shannon entropy Conditional entropy CI

1.08 ± 0.20

1.02 ± 0.19

0.390

NCI

0.73 ± 0.09

0.71 ± 0.08

0.589

Symbolic analysis (%) 0V

16.18 ± 13.13

20.68 ± 12.65

0.276

1V

45.51 ± 8.42

47.27 ± 5.87

0.448

2LV

18.65 ± 11.62

14.32 ± 8.23

0.185

2UV

19.66 ± 8.43

17.73 ± 7.83

0.458

Data are expressed as mean ± SD

The statistical analysis was carried out in Sigma Plot 11.0 for Windows. Comparisons between the groups were analyzed by the unpaired t test or the Mann–Whitney rank sum test when appropriate. Relationships between HFnu and complexity indices were verified by Pearson or Spearman correlation, depending on the data normality. Data are reported as mean ± SD, and the level of significance was set at p \ 0.05.

Results Table 1 shows the age, anthropometric and clinical characteristics, time of menopause and time of HRT of the evaluated groups. All variables were not significantly different among the groups. It is important to emphasize that a previous variance analysis (One-way ANOVA or Kruskal– Wallis One-way on Ranks, depending on the data normality) was performed by subdividing the women undergoing HRT (G2) according to the type of hormone used, showing that the type of hormone did not affect the responses of the studied variables, considering the number of volunteers of our study. So, we gathered data from subjects using hormonal therapy.

123

CI complexity index, HF high frequency in absolute units, HFnu high frequency in normalized units, HRT hormone replacement therapy, LF/HF low frequency/high frequency, LF low frequency in absolute units, LFnu low frequency in normalized units, NCI normalized complexity index, RRi RR intervals, 0V %, patterns with no variations; 1V %, patterns with one variation; 2LV %, patterns with two like variations; 2UV %, patterns with two unlike variations

The HRV spectral and symbolic analysis, as well as the Shannon and conditional entropies are shown in Table 2. HRT effect was found in the spectral analysis of cardiac autonomic modulation. LF, LFnu and LF/HF ratio were higher, while HFnu was lower in women undergoing HRT (G2) than in women not undergoing HRT (G1). Beside, RRi mean was higher between women undergoing HRT (G2) compared to women not undergoing HRT (G1). On the other hand, the RRi variance, symbolic analysis and Shannon and conditional entropies showed no significant differences. Correlations between linear analysis (HFnu) and nonlinear analysis (complexity indices) are showed in Table 3. It was observed that, in women not using HRT (G1), SE and CI presented positive correlation with HFnu, showing that the higher the vagal modulation, the higher the complexity of HRV. However, the women group undergoing HRT (G2) did not present this same trend.

Clin Auton Res Table 3 Correlations between spectral index and complexity indices Women not undergoing HRT (n = 20)

Women undergoing HRT (n = 20)

r

p value

r

p value

SE

0.523

0.018

0.380

0.099

NCI CI

0.278 0.492

0.230 0.027

0.247 0.220

0.293 0.352

HFnu

CI complexity index, HFnu high frequency in normalized units, HRT hormone replacement therapy, NCI normalized complexity index; r correlation coefficient, SE Shannon entropy

Discussion The major findings of this study were: (1) comparative analysis among the evaluated groups in the supine position showed that women not undergoing HRT presented higher CVM and lower CSM, as evidenced by spectral analysis; (2) although there was no difference in the symbolic analysis and complexity of cardiac autonomic modulation between the studied groups, it was found that the higher the vagal modulation, the higher is the complexity of HRV only in women not using HRT, whereas women group undergoing HRT did not present this same behavior as a likely result of the low value of the vagal modulation below the noise threshold. Linear method (spectral) HRV analysis showed that women who were not undergoing HRT presented higher HRV, characterized by lower CSM and higher CVM, than those who were. Liu et al. [12] evaluated HRV spectral analysis in women who were or were not undergoing HRT, with mean ages of 57 and 59 years old, respectively. They observed that women undergoing HRT presented higher vagal modulation and lower sympathetic modulation than those who were not [12]. These data conflict with the results of the present study. In addition, Neves et al. [14] compared HRV in supine and sitting position in middle-aged women who were or were not undergoing HRT, with mean ages of 53 and 56 years old, respectively. No differences were observed in cardiac autonomic modulation in the supine position and, in the sitting position, higher vagal modulation and lower sympathetic modulation were observed in women undergoing HRT than in those who were not [14]. Our data suggest that HRT users do not seem to have an advantage over nonusers regarding the autonomic modulation of HR in the supine position, only for reduction of menopausal symptoms. Christ et al. [23] evaluated HRV, blood pressure and HR at rest in the supine position of women undergoing estrogen-only HRT and women undergoing HRT with estrogen plus progestin, both groups having a mean age of 58 years old. They observed that blood pressure was not affected by

HRT, while HR was significantly higher in the combined therapy (estrogen plus progestin) group. Regarding the HRV, they found that the time domain indices representing vagal modulation (pNN50 and RMSSD) were lower in the HRT group (those taking estrogen and those taking estrogen plus progestin). However, in the subgroup analysis, they observed a significant reduction of vagal modulation compared to controls only in women undergoing combined therapy (estrogen plus progestin). They concluded that HRT could attenuate HRV in healthy postmenopausal women, and a reduction in this variable could indicate an increased risk of cardiovascular mortality. Nevertheless, these authors suggested that this effect seemed to be limited to HRT containing progestin [23]. Thus, we must consider the different types of hormones in use by our sample [conjugated equine estrogen, conjugated equine estrogen associated with medroxyprogesterone, synthetic estrogen hormone (tibolone), estradiol associated with norethisterone acetate or isoflavonoids], although statistical analysis showed that the type of hormone did not influence our results. This difference could explain why our results contradict those of studies showing that estrogen HRT promotes improved HRV characterized by higher CVM [11, 12] and lower CSM [12, 13]. Neves et al. [14], moreover, found this same HRV behavior in the sitting position with women undergoing HRT. In addition, Go¨kc¸e et al. [11] assessed HRT effects on HRV of women with mean age of 55 years old. They observed that estrogen therapy increased vagal modulation indices, while estrogen plus progestin therapy had no significant changes in HRV [11]. Furthermore, Fernandes et al. [24] reported that neither estrogen-only HRT nor HRT with estrogen plus progestin, for 3 months, promoted improvement in cardiac autonomic modulation of women with mean age of 55 and 54 years, respectively [24]. Beside, Lantto et al. [25] observed that the use of estrogen combined with medroxyprogesterone acetate for 6 months may have adverse effects on HRV in women with and without hot flashes. Also, women undergoing combined therapy may show an increased occurrence of cardiac arrhythmia compared with women undergoing estrogen only [25]. Moreover, Hautama¨ki et al. [26] reported that estrogen-only HRT causes beneficial changes in resting HR and its response to handgrip test, and these effects are partially blunted by the combined HRT (estrogen plus medroxyprogesterone acetate) [26]. On the other hand, Magri et al. [27] evaluated the effect of HRT on HRV with a 24-h Holter ECG in a group of women undergoing HRT (mean age of 54 years old) with different types of hormones (estrogens, progestin and tibolone) for 2.2 ± 2.0 years. It was observed that women undergoing HRT presented higher sympathetic modulation than those who were not, during the day, as evidenced by

123

Clin Auton Res

the LF spectral component [27]. This finding corroborates our results, which showed increased sympathetic modulation in the HRT group, as demonstrated by a significant difference in the spectral analysis LFnu index. Furthermore, Virtanen et al. [28] assessed the estrogen HRT effect on nocturnal HRV by nonlinear indices, such as approximate entropy, Poincare´ plot coefficients and fractal scaling exponents. They reported that the therapy attenuated the indices, and this may be considered a sign of a possible deleterious effect on nocturnal cardiovascular health [28]. Moreover, Fletcher and Colditz [29] showed that estrogen alone appears to be safer than combined HRT, whereas the addition of progestin may increase the risk of coronary heart disease [29]. Concerning the nonlinear analysis of HRV, it is important to emphasize that nonlinear methods—symbolic analysis and entropies—differ from traditional approaches—spectral analysis—because they consider different dynamical properties of the HR time series: linear analysis accounts for the amplitude of the HR variations, while non linear analysis is more devoted to the evaluation of the richness of the temporal scales present in the HR time course (i.e., the dynamical complexity) [3–7]. Symbolic analysis and Shannon and conditional entropies proved to be effective for assessing autonomic modulation of HR in other healthy populations [9, 10, 30, 31]. In the present study, with respect to the analysis between the evaluated groups, we observed that HRT did not affect the complexity of pattern distribution or the regularity of these patterns, which are assessed by symbolic analysis and Shannon and conditional entropies. On the other hand, correlation analysis between HFnu (linear method) and complexity indices (nonlinear methods) evidenced that, in women not undergoing HRT, the higher the vagal modulation, the higher is the complexity of HRV. However, this same behavior was not found in women group using the therapy. These findings are corroborated by studies involving pharmacological blockade and analysis of HRV complexity, which showed that a reduction in vagal modulation leads to decrease in complexity indices. According to Porta et al. [32, 33], the vagal blockade induced by high dose of atropine leaves the cardiac pacemaker under the sole sympathetic control, thus reducing the complexity of the cardiac control. Moreover, Porta et al. [33, 34] reported that the sympathetic blockade obtained via administration of propranolol did not change the complexity, suggesting that the contribution of sympathetic modulation to the complexity is not significant. Thus, the findings of this study suggest that women not undergoing HRT presented favorable effects on cardiac autonomic modulation, characterized by higher vagal modulation and lower sympathetic modulation, analyzed by spectral analysis. In addition, complexity analysis showed that only women not using HRT presented positive

123

relationship between vagal modulation and complexity of HRV. Since a significant correlation is expected, the absence of this association in women undergoing HRT suggests that in this group vagal modulation drops below a minimal value preventing the relation to become apparent in presence of measurement noise. It could be hypothesized that this minimal value of vagal modulation is less favorable than the level of vagal modulation in women not using the therapy. Thereby, considering the viewpoint of autonomic nervous system control, both spectral and complexity analyses showed that the HRT, independently of the hormone type, seems not to be advantageous for the cardiac autonomic modulation in the age range studied. Therefore, we must consider that an association of linear and nonlinear methods for HRV analysis is important for more complete understanding regarding the cardiac autonomic modulation in different situations. Thus, although HRT reduces menopausal symptoms (hot flashes, cold flashes, irritability, mood swings and depression) improving quality of life, it appears not to have beneficial effects on HRV. A possible compensation for HRT’s effect on cardiac autonomic modulation would be aerobic exercise, which promotes increased vagal modulation and improved HRV [35]. This study presented some limiting factors, such as the lack of follicle stimulant hormone (FSH) dosage as a way to evaluate the menopause. However, the postmenopausal phase was clinically ensured by amenorrhea for at least 1 year. It is emphasized that the volunteers studied had a mean duration of menopause around 10 years. Another limitation is the absence of a group receiving a placebo hormone. Nevertheless, the aim of this study was to evaluate the differences in the cardiac autonomic modulation between a women group using the HRT and another group not using, and not to evaluate the efficacy of the hormone. Furthermore, another limitation of this study is nonhomogeneity of the HRT group regarding the type of hormone, but we have to consider that women who underwent HRT in the general population use different types of hormone, which leads to the difficulty to screen a homogenous group. Moreover, a previous analysis performed in this study, showed that the type of hormone did not influence the indices of HRV, which allowed evaluating a single group in use of hormone therapy, we call G2. Based on our results, we conclude that women who did not use HRT apparently have higher CVM and lower CSM in terms of quantitative properties assessed by linear analysis. In addition, the positive relationship between CVM and complexity of HRV suggests that women not undergoing HRT present cardiac autonomic control more favorable compared to women using the therapy, showing that HRT is not advantageous for cardiac autonomic modulation in the evaluated groups, independently of the hormone type.

Clin Auton Res Acknowledgments Research supported by Fundac¸a˜o de Amparo a` Pesquisa do Estado de Sa˜o Paulo—FAPESP (#2010/52703-7 to Nata´lia M. Perseguini and #2010/52070-4 to Aparecida M. Catai). 14. Conflict of interest of interest.

The authors declare that they have no conflict

References 1. Task Force of the European Society of Cardiology and the North American Society of Pacing and Electrophysiology (1996) Heart rate variability: standards of measurement, physiological interpretation and clinical use. Circulation 93:1043–1065 2. Bigger JT Jr, Fleiss JL, Steinman RC, Rolnitzky LM, Kleiger RE, Rottman JN (1992) Frequency domain measures of heart period variability and mortality after myocardial infarction. Circulation 85:164–171. doi:10.1161/01.CIR.85.1.164 3. Huikuri HV, Makikallio T, Airaksinen KE, Mitrani R, Castellanos A, Myerburg RJ (1999) Measurement of heart rate variability: a clinical tool or a research toy? J Am Coll Cardiol 34:1878–1883. doi:10.1016/S0735-1097(99)00468-4 4. Huikuri HV, Makikallio TH, Perkiomaki J (2003) Measurement of heart rate variability by methods based on nonlinear dynamics. J Electrocardiol 36(Suppl):95–99. doi:10.1016/j.jelectrocard. 2003.09.021 5. Jokinen V, Tapanainen JM, Seppanen T, Huikuri HV (2003) Temporal changes and prognostic significance of measures of heart rate dynamics after acute myocardial infarction in the betablocking era. Am J Cardiol 92:907–912. doi:10.1016/S00029149(03)00968-8 6. Makikallio TH, Huikuri HV, Hintze U, Videbaek J, Mitrani RD, Castellanos A, Myerburg RJ, Moller M (2001) Fractal analysis and time- and frequency-domain measures of heart rate variability as predictors of mortality in patients with heart failure. Am J Cardiol 87:178–182. doi:10.1016/S0002-9149(00)01312-6 7. Makikallio TH, Tapanainen JM, Tulppo MP, Huikuri HV (2002) Clinical applicability of heart rate variability analysis by methods based on nonlinear dynamics. Card Electrophysiol Rev 6:250–255 8. Porta A, Guzzetti S, Montano N, Furlan R, Pagani M, Malliani A, Cerutti S (2001) Entropy, entropy rate, and pattern classification as tools to typify complexity in short heart period variability series. IEEE Trans Biomed Eng 48:1282–1291. doi:10.1109/10. 959324 9. Guzzetti S, Borroni E, Garbelli PE, Ceriani E, Della Bella P, Montano N, Cogliati C, Somers VK, Malliani A, Porta A (2005) Symbolic dynamics of heart rate variability: a probe to investigate cardiac autonomic modulation. Circulation 112:465–470. doi:10.1161/CIRCULATIONAHA.104.518449 10. Porta A, Tobaldini E, Guzzetti S, Furlan R, Montano N, GnecchiRuscone T (2007) Assessment of cardiac autonomic modulation during graded head-up tilt by symbolic analysis of heart rate variability. Am J Physiol Heart Circ Physiol 293:H702–H708. doi:10.1152/ajpheart.00006.2007 ¨ rem C, Erdo¨l C, O ¨ zdemir S 11. Go¨kc¸e M, Burhan K, Yilmaz R, O (2005) Long term effects of hormone replacement therapy on heart rate variability, QT interval, QT dispersion and frequencies of arrhythmia. Int J Cardiol 99(3):373–379. doi:10.1016/j.ijcard. 2003.03.030 12. Liu CC, Kuo TBJ, Yang CCH (2003) Effects of estrogen on gender-related autonomic differences in humans. Am J Physiol Heart Circ Physiol 285:H2188–H2193. doi:10.1152/ajpheart. 00256.2003 13. Rosano GMC, Patrizi R, Leonardo F, Ponikowski P, Collins P, Sarrel PM, Chierchia SL (1997) Effect of estrogen replacement

15.

16.

17.

18.

19.

20.

21.

22.

23.

24.

25.

26.

27.

therapy on heart rate variability and heart rate in healthy postmenopausal women. Am J Cardiol 80(6):815–817. doi:10.1016/ S0002-9149(97)00528-6 Neves VFC, Silva de Sa´ MF, Gallo L Jr, Catai AM, Martins LE, Cresceˆncio JC, Perpe´tuo NM, Silva E (2007) Autonomic modulation of heart rate of young and postmenopausal women undergoing estrogen therapy. Braz J Med Biol Res 40(4): 491–499. doi:10.1590/S0100-879X2006005000080 De Meersman RE, Zion AS, Giardina EGV, Weir JP, Lieberman JS, Downey JA (1998) Estrogen replacement, vascular distensibility, and blood pressures in postmenopausal women. Am J Physiol Heart Circ Physiol 274:H1539–H1544 Vongpatanasin W, Tuncel M, Mansour Y, Arbique D, Victor RG (2001) Transdermal estrogen replacement therapy decreases sympathetic activity in postmenopausal women. Circulation 103:2903–2908. doi:10.1161/01.CIR.103.24.2903 Huikuri HV, Pikkujamsa SM, Airaksinen KE, Ika¨heimo MJ, Rantala AO, Kauma H, Lilja M, Kesa¨niemi YA (1996) Sexrelated differences in autonomic modulation of heart rate in middle-aged subjects. Circulation 94(2):122–125. doi:10.1161/ 01.CIR.94.2.122 Ross RK, Paganini HA, Mack TM, Henderson BE (1989) Cardiovascular benefits of estrogen replacement therapy. Am J Obstet Gynecol 160:1301–1306 Adams MR, Kaplan JR, Manuck SB, Koritnik DR, Parks JS, Wolfe MS, Clarkson TB (1990) Inhibition of coronary artery atherosclerosis by 17-beta estradiol in ovariectomized monkeys. Lack of an effect of added progesterone. Arteriosclerosis 10(6): 1051–1057. doi:10.1161/01.ATV.10.6.1051 Malliani A, Pagani M, Lombardi F, Cerutti S (1991) Cardiovascular neural regulation explored in the frequency domain. Circulation 84:482–492. doi:10.1161/01.CIR.84.2.482 Pagani M, Lombardi F, Guzzetti S, Rimoldi O, Furlan R, Pizzinelli P, Sandrone G, Malfatto G, Dell’Orto S, Piccaluga E (1986) Power spectral analysis of heart rate and arterial pressure variabilities as a marker of sympatho-vagal interaction in man and conscious dog. Circ Res 59(2):178–193. doi:10.1161/01.RES.59. 2.178 Porta A, Faes L, Mase´ M, D’Addio G, Pinna GD, Maestri R, Montano N, Furlan R, Guzzetti S, Nollo G, Malliani A (2007) An integrated approach based on uniform quantization for the evaluation of complexity of short-term heart period variability: application to 24 h Holter recordings in healthy and heart failure humans. Chaos 17(1):015117. doi:10.1063/1.2404630 Christ M, Seyffart K, Wehling M (1999) Attenuation of heart-rate variability in postmenopausal women on progestin-containing hormone replacement therapy. Lancet 353:1939–1940. doi:10. 1016/S0140-6736(99)00639-X Fernandes EO, Moraes RS, Ferlin EL, Wender MCO, Ribeiro JP (2005) Hormone replacement therapy does not affect the 24-hour heart rate variability in postmenopausal women: results of a randomized, placebo-controlled trial with two regimens. Pacing Clin Electrophysiol 28:S172–S177. doi:10.1111/j.1540-8159. 2005.00041.x Lantto H, Haapalahti P, Tuomikoski P, Viitasalo M, Va¨a¨na¨nen H, Sovija¨rvi AR, Ylikorkala O, Mikkola TS (2012) Vasomotor hot flashes and heart rate variability: a placebo-controlled trial of postmenopausal hormone therapy. Menopause 19(1):82–88. doi:10.1097/gme.0b013e318221bae8 Hautama¨ki H, Haapalahti P, Piirila¨ P, Tuomikoski P, Sovija¨rvi A, Ylikorkala O, Mikkola TS (2012) Effect of hot flushes on cardiovascular autonomic responsiveness: a randomized controlled trial on hormone therapy. Maturitas 72(3):243–248. doi:10.1016/ j.maturitas.2012.04.001 Magri F, Gabellieri E, Busconi L, Guazzoni V, Cravello L, Valdes V, Sorrentino AR, Chytiris S, Ferrari E (2006)

123

Clin Auton Res

28.

29.

30.

31.

Cardiovascular, anthropometric and neurocognitive features of healthy postmenopausal women: effects of hormone replacement therapy. Life Sci 78:2625–2632. doi:10.1016/j.lfs.2005.10.036 Virtanen I, Ekholm E, Polo-Kantola P, Hiekkanen H, Huikuri H (2008) Postmenopausal estrogen therapy modulates nocturnal nonlinear heart rate dynamics. Menopause 15:693–697. doi:10. 1097/gme.0b013e31815dba86 Fletcher SW, Colditz GA (2002) Failure of estrogen plus progestin therapy for prevention. JAMA 288:366–368. doi:10.1001/ jama.288.3.366 Takahashi ACM, Porta A, Melo RC, Quite´rio RJ, da Silva E, Borghi-Silva A, Tobaldini E, Montano N, Catai AM (2012) Aging reduces complexity of heart rate variability assessed by conditional entropy and symbolic analysis. Intern Emerg Med 7(3):229–235. doi:10.1007/s11739-011-0512-z Perseguini NM, Takahashi AC, Rebelatto JR, Silva E, Borghi-Silva A, Porta A, Montano N, Catai AM (2011) Spectral and symbolic analysis of the effect of gender and postural change on cardiac autonomic modulation in healthy elderly subjects. Braz J Med Biol Res 44(1):29–37. doi:10.1590/S0100-879X2010007500137

123

32. Porta A, Guzzetti S, Furlan R, Gnecchi-Ruscone T, Montano N, Malliani A (2007) Complexity and non linearity in short-term heart period variability: comparison of methods based on local non linear prediction. IEEE Trans Biomed Eng 54:94–106. doi:10.1109/TBME.2006.883789 33. Porta A, Castiglioni P, Bari V, Bassani T, Marchi A, Cividjian A, Quintin L, Di Rienzo M (2013) K-nearest-neighbor conditional entropy approach for the assessment of the short-term complexity of cardiovascular control. Physiol Meas 34:17–33. doi:10.1088/ 0967-3334/34/1/17 34. Porta A, Castiglioni P, di Rienzo M, Bari V, Bassani T, Marchi A, Takahashi ACM, Tobaldini E, Montano N, Catai AM, Barbic F, Furlan R, Cividjian A, Quintin L (2012) Short-term complexity indexes of heart period and systolic arterial pressure variabilities provide complementary information. J Appl Physiol 113:1810– 1820. doi:10.1152/japplphysiol.00755.2012 35. Albinet CT, Boucard G, Bouquet CA, Audifren M (2010) Increased heart rate variability and executive performance after aerobic training in the elderly. Eur J Appl Physiol 109:617–624. doi:10.1007/s00421-010-1393-y